S4 E11 - John Dues - Transforming Education Through Deming's Principles

In this episode of the Profound Podcast, I'm joined by John Dues, Chief Learning Officer at United Schools Network in Ohio and an innovative educator who has ingeniously applied W. Edwards Deming's theories to the educational sector. John shares his journey of discovery into Deming's methodologies and how these have profoundly influenced his approach to improving educational systems.

Our discussion delves into John's initial skepticism and eventual embrace of Deming's System of Profound Knowledge, which consists of understanding systems, variation, psychology, and knowledge theory. John explains his experiences in applying these principles within his school network, particularly how they've been used to enhance operational efficiency and educational outcomes amidst the challenges posed by remote learning during the pandemic.

The conversation also touches on John's approach to leadership and learning, emphasizing the necessity of continuous improvement and systems thinking in educational administration. He highlights the importance of aligning educational strategies with Deming’s philosophies to foster environments that nurture rather than stifle, promoting an organizational culture where insights into systemic improvements are encouraged and valued.

John's innovative application of Deming's theories offers a template for transformative leadership in education, demonstrating the relevance of these age-old concepts in addressing modern challenges within the sector.

You can find John Dues' LinkedIn Below:
https://www.linkedin.com/in/johnadues/

Transcript

John Willis: [00:00:00] Hey, this is John again and another profound podcast and, you know, sort of mostly all things Deming. So, today is definitely Deming, Deming Treat. So we got an author. I think there's only been two books recently come out about Deming and, and and I think John, do you want to introduce? I think yours is one of them and mine's the other.

So, yeah. Yeah, 

John Dues: I think so. You want me to introduce the book? 

John Willis: No. Yeah, no, it's good. You tell, tell us who you are and, and you know, and then we can talk about the book. Okay, all sorts of good stuff. 

John Dues: Yeah, sounds good. Thanks for having me. My name is John Dues. I professionally, I'm the chief learning officer for United schools network here in Plums, Ohio.

So we're just we're a small school system. Basically, we have 4 buildings and I help a lot on the operational front continual improvement front. And I've been a principal in our [00:01:00] network and, and some other things, but, you know handful of years ago sort of came across Deming and more and more have started using Deming in my work to the point where I wrote about the, the, those experiences in a book, my sort of learning process and learning how to apply the lessons of Deming to the education system and kind of in that way.

You know, going deeper and learning myself. So yeah, 

John Willis: oh, okay. No, we got a lot to go over because your book was excellent. The I did get around to reading it. You know, these days I try to do them first on audio. So we're going to get you an audio man. 

John Dues: I know. Same, same. And I've read your book too. So 

John Willis: cool.

So I think, you know, the got a lot of questions, but I guess the one we, you know, we, we, you, you, you coming on to the the original profound book club that the, the guys kept running and they did your books next. And, and I had [00:02:00] asked you a question in the book club, which I like to ask people, like how Deming, what Deming, what was the sort of thing that got you involved in, in Dr.

Deming? 

John Dues: Yeah, that's a really good question. So I you know, I've informally always, you know, tried to improve in my work. Then about 2015, late 2015, I got an email and it was from an education policy group and they highlight various books among other things. And one, one of the books was called Learning to Improve How American Schools Can Get Better at Getting Better, written by this Team led by Tony Brake at the Carnegie Foundation for the Advancement of Teaching.

Their whole emphasis was on a methodology called Improvement Science and they were trying to bring that from healthcare and other places into schools basically and they were learning how to do that. So I, the book piqued my interest. I grabbed it, read it, started going to the Carnegie Improvement Summits and learned.

You know, what they call improvement science methods and tools and [00:03:00] a few times over the next, you know, four or five years, I had come across Deming's names in the various books I was reading and you know, decided to check it out. Probably about 2018, I'd read his name a number of times and said, let me, let me check who this person is out.

So I go to the Deming Institute website. I go to the system of Profound Knowledge page, which I had never been exposed to, and sort of read the four components, and I had two sort of reactions. One was, well, who calls their stuff their own theory found It seems egotistical. And my second reaction is when I look at these four components, you know, systems and variations, psychology and knowledge, it's basically like incomprehensible.

I didn't get it. At all, so I sort of went away from that page. I remember is like a Saturday morning. 

John Willis: You mentioned this. Why? What made it incomprehensible? It was the combination of the four or I had a reaction, but it was a little different. But yeah, 

John Dues: [00:04:00] I think so. I mean, I think, you know, I had sort of. You know, I was a psychology major undergrad.

So some of the psychology stuff was familiar. The theory of knowledge and the variation stuff in particular, I had never sort of solve variation or data frame in that way. And so. I didn't have much of an understanding of what he was even talking about with theory of knowledge. Systems thinking, I had sort of read some things on that front, so that was a little more familiar.

But yeah, it was variation in knowledge, theory of knowledge, that frame, plus all four of those things, how are they supposed to work together. And so I sort of went away from the institute website, but I, you know, would keep coming across his name. And I finally went back to the website like two years later, and I don't know what had changed.

Maybe I'd learned more or something more exposure to the improvement science stuff and something just clicked. Not that I understood the theory or anything like that, but it was, it lit a fire and I just started reading. Book after book after book, some that [00:05:00] he wrote and then some that others wrote about his work or, you know, related work and it started sort of slowly coming into focus over the next six months and to the point where you know, I, I decided I'm going to sit down and started writing a book about Deming in about September of 2020.

I'd also watched some videos and I realized that videos of him speaking or giving workshops or talks or whatever, and I also realized like, you know, he's a pretty humble guy and it's not that he's saying he's profound. It's the knowledge you get about your organization through this lens that's profound.

So I learned a few things along the way, I guess. 

John Willis: You know, it's funny. I didn't have that reaction about profound and I'm not sure why, because I think I was already eased into this whole, you know, I, I came in through Goldratt and then by that time I was like, okay, these guys are different. So if he calls me, I'm fine.

But I, I do recall people coming to me and I try to explain it like. Profound, [00:06:00] really? Is that what he calls it? You know, and, but you know, I, there's a part that I had in my book, right, and he was being interviewed years after Japan, and he said something, and I, I, the quote isn't exactly, but he said there was only one person with profound knowledge in Japan, and, and then you could see how Deming was so particular about his words.

So one way you could have take that is I was only the one who was smart enough to teach him, but that's not what he meant. Yeah. There were this, there was this thing of profound knowledge and nobody was teaching all those four components. So, yeah. So yeah, you, you sort of get over that really quick about like, yep.

It's not ego, it's profound. 

John Dues: Right, right. Yeah. Absolutely. 

John Willis: One of the things I was, I really enjoyed in your book, right, which I, you know, I've read quite a few, you know, you know, I, you know, I, you know, I, I started my journey started a little over 10 years ago with, with Deming, and I've documented that and, and, and.

And I, you know, I paid attention, found some stories, but [00:07:00] during the pandemic, I decided, you know, like, if you're ever going to write this book, now's the time, or I was just shut up about it. Right. And so then I just started ordering every book there was right. And I, I gotta say, I've, I've been a big fan of systems thinking I I've read, you know, fair amount of stuff with Manila Meadows and Jay Forrester.

People, I'm not an expert by any means. But I didn't see in a lot of the books. that I had read. Maybe I wasn't reading it right. Maybe I wasn't sort of paying attention. A lot of emphasis on systems thinking. You right out of the gate. make it really clear how important that is. You know, I mean, I know it is appreciative system and like, yes, of course, but, but you're, you really, I think you did a really good job of emphasizing system thinking, you know you know, I, I've done a lot of podcasts with Bill Bellows and he's fantastic at this, right?

So I, I was wondering, you said you already had a background in systems thinking or. [00:08:00] 

John Dues: Not, not formally. I mean, I think it was really, I mean, well, it was very informal. I think, I mean, one working in a school system, although ironically, a lot of school systems don't really operate like, like systems and, you know, but I had been in this sort of startup organization where we went from, you know, one school with a single grade to a network of four schools.

And so, yeah. We had to build a system. And so why hadn't I didn't have the formal training or language, you know, I think we did some things that actually, you know, lined up with appreciation for a system for whatever reason. Probably because we started small and I like the people that then filtered out and did some of the growth stuff.

And so I think that helped. In terms of the mindset and the framework that then lined up with some of the principles. But yeah, no formal, no formal training other than what I have. 

John Willis: Yeah, that's interesting. I've done a fair amount of startups and I've never really thought of that. But the way, like, starting with something small and [00:09:00] watching it grow.

Has so many like so much scar tissue in negative and positive ways, right? And one of the ways is oh, you know, it was so different when it was the five of us or the ten of us You know, and you bring in a marketing team and like that's different and like, yeah I I really thought about that is that's a good way to see Systems, you know, yeah.

Yeah, that's that's pretty cool. Yeah, i'm tempted to ask you, you know, like I think I you know, I guess you could say that like SOPK, system profound knowledge, can never be explained enough, right? So but like, I think you did such a good job of mapping it to real world scenarios in the school system.

I mean, if you had to give, like, if people are popping in or, you know you know, at this point you, you know, if you've listened to this podcast at this point, you sort of know what system profound knowledge is, but I think hearing your, what, what is like, give me the elevated pitch of system profound knowledge based on your experience with [00:10:00] it.

John Dues: Yeah. So the way I think about it is I usually call it a management philosophy or a management theory. And I think it's helpful. So when you think about system of profound knowledge, there's the system part that then connects to the profound knowledge part. And so when I, my superficial understanding was profound was like, Deming saying his ideas were profound, but what, you know, you dig into that profound knowledge is really this deep understanding, this unique lens that you see your organization through when you connect the dots between systems and variation theory of knowledge.

In psychology, that's so that's one part. And then the system part is those four things in and of themselves are this interlinked system. So you can't just be good at one. You have to sort of take all four of them together and put these things together. And so when you [00:11:00] start talking about it, you start, you start realizing Like why they are mutually supporting, I guess is what I would say is.

So, you know, if you are going to, you know, one of the, one of the charts in the book, I chart the, The remote learning engagement of students in an eighth grade math class during the pandemic. And so it's like, we define what engagement and then we measured it every day. How many, what percent of the kids were engaged by our definition.

And so, if I'm a leader of, let's say, I'm the principal of that building, and the engagement level is not where I want it to be, I may be going to that math teacher and saying, what's going on here? Like, why can't you do better than this? But with this Deming lens, you know, with the psychology part, with the understanding variation part, I'm going to them and I'm saying, okay, you know, over the first few weeks or month of the school year, we're actually starting to understand the capability of this remote learning system.

You [00:12:00] know, we're all trying to figure this out together. We see this data unfold over the first month, you know, we're bouncing around about two thirds of the kids are engaged on any given day. That's not, you know. It's not a pat on the back to the teacher, and it's also not a reprimand. It's just what it is, what is what the system.

And so that's a completely different mindset, you know, what am I going to do as the school leader, how can I work with the teacher, how can I work with the teacher and the students to improve, you know, in this case, their engagement level. So it's a completely different lens. With which to, you know, view data and then to interact with the people that are going to have to improve the systems that's that are producing the data.

So that's, you know, that's kind of how I think about it and why all four of the parts together are so, so important 

John Willis: for yeah, good job of there's some things I guess I, you know, now the risk of asking you education questions just because I'm interested, but you know, [00:13:00] my, my my youngest son Was basically going through 11th and 12th grade in during the pandemic, right?

And and you know, he is my oldest son was a straight A student since he was like, you know, five years old But my son was like me, you know, like we're squirreled. That's a question Oh, we get distracted really easy, but but he's a good B solid soon any you know, he survived it You know, but he said dad, you know the Some of the kids just, you know, they didn't know how to manage anything and they just got, you know, like in his, you know, sort of uneducated, non educated version of what happened to most of the kids in his grade group just got, you know, you know, like was a terrible, You know, for Noah, you know, like for all the variables that creates it.

I guess the question I'm trying to ask then is Did, were you able to use Deming's principles to actually affect improvement because you were looking at the data during the process? 

John Dues: That's a really good question. I mean [00:14:00] I think it's hard to say for sure. We were also, you know, when I went back to the Deming Institute website that second time and things started to click was You Literally February 2020.

Okay, a month before the schools shut down. But, you know, in a lot of ways. Yes, because so here's here's a good example. So I talked about remote learning engagement. So 1 of our principals. Had a great, you know, structure. She had the spreadsheet and all the teachers would put in their engagement levels for each day.

And they would compare them. And they, they, they even added this green, yellow, red kelp cutter, color coding system, this was a middle school. And then our other middle school was doing something completely different. And so we said, well, okay. At the middle school, that's collecting the data. I just read some Deming.

He talks about operational definition. So as a starting point, [00:15:00] what is engagement, right? And, you know, even at the school where they collect the data, it was kind of like, well, you know, the kids are doing most of the work. I said, is it most of the work? Was it all of the work? Can they miss one assignment a week?

And then the other school had a completely different definition of what it meant to be engaged. Just something as simple as that. So what we did was define it as a group. What does it mean to be engaged, as we're using that word in a remote learning environment. And then, I began to take the data out of that spreadsheet and put it in a process behavior chart.

And so basically I was sort of learning this as I was doing this and saying, you know, you know, you, you've assigned these sort of stoplight colors to this engagement data. Green is above 80%, yellow is in the 70s, red is 60s, 50s, whatever I said, but that's like completely arbitrary. Right. Look at this data in this other way.

See how it's just moving up and down. [00:16:00] And she was like, Whoa, okay. Right. And then with one of the grade levels, what we started to do was say, okay, we've, we've mapped this on a process behavior chart. We want to make this better. So what we, what can we do as a plan to study act cycle? And we actually drew a line in the chart.

It's that we're going to start a process or I'm sorry, a PSA cycle right here. And we're going to try something different to improve the engagement levels. And I would actually get on and watch the lessons. And then we, we planned some activities where we gathered some feedback from students on zoom.

Like, you know, why, why some days you engage some days you don't, some days you can complete your stuff. Some days you don't, we did some surveys. Both like one on one in Zoom rooms and then also like, you know, just send them a Google form and they, you know, fill out and give them some ideas. And we took some of that and designed some PDSAs.

I mean, I can't say, you know, that we improved things significantly, but in a lot of [00:17:00] ways, you know, we started applying the lessons like right away. 

John Willis: Yeah. A lot to learn during a very unique situation. I think, I love that you cover operational definition. You know, for me. It's funny, I, like you know, like, you sort of, like, for me, you get confused, right?

Is this all about variation? Oh, it is systems thinking. What's the psychology thing? You know, and then, you know, sort of like, if you read my book, I really had to sort of spend a lot of time trying to figure out what this theory of knowledge really was, right? That unravels very deep. But it seemed to me like it was very late in my book where somehow the light bulb went off about operational definitions.

Right. It's almost like I, if I, if, if, like, I was around when Deming was writing New Economics and I forget which one he, he, he creates the most principles around it, but, like, I would want to say, dude, like, you need to tell me, like, this is a critical principle. Before I read this book. And and, and, [00:18:00] and, and so I, when I went back, I was like, Oh my God, that connects everything.

It seems like you've got it right away though. 

John Dues: Maybe there was a little bit of the, you know, because with the psychology major, you didn't have to do, I mean, we did use that terminology, even in undergrad operational definitions for things that you're going to experiment on, so that was important to very carefully define it.

So maybe I had a little bit of background in that, but then sort of, I remembered, you know, 20, 20 years later. But I do think a lot of Deming stuff, I mean, some of it is technical and hard to understand, but a lot of it is low hanging fruit, and I think that's, operational definitions is right up there, because You know, so much of the stuff that we talk about, that we're so certain about, is built on a house of cards.

You ask one or two questions and the whole thing starts to fall apart. Like, and operational definition is a good example. Like, going to that principle, no fault of our own, what do you mean by engagement? How do you define that, right? You know, a lot of times people have this rough idea, but they haven't actually written this down in a way that [00:19:00] maps to the, to the numbers that they're gathering from their, from their system, so.

Yeah, 

John Willis: no, operation differences are like this thing you can't unsee once you get it, right? Yeah. In my field, I, you know, once I really understood it, I, I'd always had some problems with the way people measured the way we deliver software. We do this, it takes this amount of time to hear to hear. And, and then like, I always felt, then I, when I understood like the, went back to red Wheeler stuff, right.

And really understood it much better than I could go into a client and say, okay. Like, when does it start? And when does it end? Oh, wow. Then you talk to another team. Well, we don't know another team. Oh, it's when, you know, like, yeah. And then you start seeing, like, it is that not being able to unsee this, like almost every question was got to have clear operational definitions for really to glue any of this together.

So, 

John Dues: yeah, I think that's cool. Another piece of that and to connect it to the psychology component again, is You know, when you build a measurement system like we're doing some strategic planning and so I'm sitting [00:20:00] with different sort of department heads and saying, okay, what's most important to you to know what measures are going to give you some you know you know, the, the look to say that, you know, things are going well in your department.

And so when you sit and you start to. interrogate, okay, why did you pick this thing over this thing? Or you know, then what about this other metric? Isn't this really important? And, you know, you start to hear things like, well, you know, that metric, if we do that, that's not going to make my department look really good.

And I'm like, this isn't accountability, this is improvement. And so if I'm sort of, even in a place that I think. You know, people can speak up in our organization, but even in a place like that, you know, there is fear there, whether you admit it or not, no matter how, you know, great your organization is, there is fear there and people are productive of their area or whatever.

And so you like, when Deming says drive out fear, you like really have to drive out fear. That's why the rating and [00:21:00] ranking has to be removed completely. Right. 

John Willis: No, I want to come back to that too. But the, the other thing about PDSA that I, I liked that you covered, I, you know, I, it, it, it's one of those things that seems obvious, but I hadn't really thought about it.

I think a lot about, you know, scientific method and deductive reasoning or abductive reasoning. But I, I love the way you sort of like had made me sort of step back for a second. I think that the PD DSA is really this, this going between deductive and inductive and de mm-Hmm. That was pretty, I thought it was pretty clever.

And I hadn't really thought of it that way. 

John Dues: Yeah, I think I saw some version of that diagram or that representation in the improvement guide from from the Institute for Healthcare Improvement. So when I saw that, it just really clicked that, you know, plan and then do you're going from inductive to deductive.

 That made a lot of sense when I saw it. Yeah, deductive, 

John Willis: inductive, yeah, yeah. Yeah, sorry, 

John Dues: I did it, yeah. 

John Willis: I, it's, you know, like, those are the ones you like, I, like, I don't make this mistake, but I [00:22:00] remember when I was younger, horizontal and vertical, like, Ah! Right, right, right. Inductive, deductive is like that too.

No, no, but it was, it was kind of cool, because it really is that sort of you know, sort of deductive process, and then you sort of, the results are, okay, now what am I doing with the results, and then I, I loop that. So I thought that was really cool. 

John Dues: Yeah, appreciate that. 

John Willis: The other thing I think you know, so I, like, I, I think one of the things that so in IT and software delivery, and you're starting to meet some of the people that, you know, that sort of have been, I work with and great people, yeah, struggle to figure out how to implement the, you know, the sort of the, understanding variation, right? Like, it's not used, I would say, because part of this is, I think you do a great job explaining, and I'll sort of ask you to explain it to this audience, the difference between enumerated and analytical statistics. But, but you do a really good job of, you have real data, you know, you attack the problem with like, okay, I understand this guy.

I got tons of data. Let me see how [00:23:00] this works. So I think I like, like that, you know, you had, like, you were able to say, like, like, this is variation, this is common cause, this is special cause, but this is what we did. And, and so I think in, in our field, just there aren't a lot of people that, there are starting to be these young people that have read my book and sort of read your book and, and really experiment with Demi's ideas, but we don't, I don't see a lot of examples like we do in healthcare in education and other places.

And I thought like, you know, like you were able to sort of implement a lot of learning from really from like the context specifics of data in your book. 

John Dues: Yeah. I mean, I think, well, one, you know speaking of like driving out fear, like, you know, there's going to be a lot of leeway and it's sort of. start thinking about how to apply these methods right in our, in our network.

So I would sort of, you know, you mentioned Wheeler, [00:24:00] he's sort of my go to guy for the understanding variation stuff. Mark Graben too, because I think his book Measures of Success makes the Wheeler stuff. easier to understand. He's got nice color coded process behavior charts. So I had the freedom to start practicing this stuff.

And I would, I had enough sort of trusted colleagues that I could put it in front of them and say, does this make any sense to you? You know, what, what would I need to do to explain this to teachers? And we're still like trying to figure that out because the process behavior charts are, you know, relatively new, but I had a Gates grant and I designed a fellowship where we.

Accepted people internally to sort of go through a process where they learn these methods. And so I learned sort of what was working with the people that are brand new to it, even newer than me. And then at the same time, we got a, another grant from Dell and one of my colleagues who I've worked pretty closely with Deming stuff on, and he knew this probably about as good as I [00:25:00] did.

He worked with some folks, some developers to create dashboards that actually used. Process behavior charts, which the Dell people that we were working with had never seen before this. And they thought that was actually pretty cool. And they taught us how to do dashboards and data, you know, using Google data studio or whatever.

And then we taught them about process behavior charts. So there's this back and forth. I don't know if I answered your question there, but that's sort of, I think you did the grants, 

John Willis: the experimentation, the freedom. I think that that's, I think that probably is a missing piece. And a lot of people in my area, it's like, it's not that you don't have the freedom.

 But you're sort of, your plate is so full anyway that, you know, like, and you get another matrix. Sorry, one, 

John Dues: one part of that. So that was key was, and this was part of the study. I, you know, I started listening to the old podcast Institute website. And basically every time they did an interview, I would go look the person up and then I would try to contact them.

And like, the [00:26:00] 1st person that I contacted, I think was Kelly Allen is involved with the Institute. And he said, well, if you're an educator, you should talk to David Langford just. emailed him and he called me right away and we had a conversation and what I ended up doing is using some of that Gates grant to work with David.

Oh wow. Sort of teach me and coach the people that were in the fellowship. So the first cohort, I didn't know as much. David actually did the coaching and I would watch as he coached the fellows in these methods. And so I'm saying all that to say is like, if you can reach out to someone that's. It's a few steps or many steps ahead of you.

I was lucky enough to, you know, David's obviously one of the most preeminent people in the education sector that uses these methods. I was able to work with him for 15 weeks or so and got a crash course. 

John Willis: I think the thing in our field is that there is really no preeminent expert right like that.

Nobody's really, at least. Yeah. You know, my view, I hadn't seen anybody really do a lot of sort [00:27:00] of infrastructure operations. It I guess this is gonna be a could be a terrible question. It would be the best question of the podcast. Like , like if, if you had to think about what were the sort of, you know, if you had to identify what, what were some real aha moments about using, you know, control CROs you know, statistical process control, like things that like.

My goodness, like I never expected we could have seen that 

John Dues: well. So in in Wheeler's book, Understanding Variation, there were two sections that really stood out. One was like a within a chapter of a subtitle numerical naivete. Oh, 

John Willis: I love that. I was going to ask you about that. I love that. 

John Dues: And I used that sort of characterization in my book.

And the second one was writing fiction. And then when I understood what he was saying, we're sort of naive, how we typically look at data. We don't take into account how data is moving up and down naturally. So we don't really understand what's happening with our data. And then. What we [00:28:00] often do is sort of write a story about our data.

That's not, not in the fairest way or cheating there there. Cause there's been data scandals in education. School systems have gotten in trouble, faking data or whatever. That's not what I'm talking about. I'm talking about like. Dressing it up for some audience parents or funders or whoever, you know, focus on this day, but not this date over here, or, you know, you stop arbitrarily 2 years ago, because if you go back 3 years, it really doesn't look that good.

You do all these things and you just really don't understand. And so once I read that book. I started seeing this everywhere. I turned everywhere. I turned. There's a hard Department of Education document. It was like test scores, state test scores. And it said Ohio students improved another year or whatever.

And I just looked at the data. I'm like, there is just zero evidence you map this out. And this data is literally, you know, [00:29:00] 120, 000 or so Ohio students take each test. Grade level test each year. So there's a big sample size. And so the data literally just goes like this. Up a little bit, down a little bit, up, down.

And there's this average running through it. And we call that improvement. Or we really try to manipulate where it says, two years ago there was improvement, the last year there wasn't, and that's okay. Because two years from two years ago, you know, it's just like arbitrarily choosing these things and dancing around and spending all, we waste so much time doing that and not actually improving anything.

So that was the biggest thing. It's just, Having a way to understand, okay, when has improvement occurred? When hasn't it? There's these rules, you know, you can use in process behavior charts to make that distinction. 

John Willis: Yeah, 

John Dues: you know, even just even if you have a process behavior chart and just when you're going to try an intervention, just draw that vertical line and label it PSA and a simple annotation like [00:30:00] with that remote learning example, maybe we'll try shorter lesson videos.

And we're going to PDSA it. If things get better, I can actually tie the improvement back to something and I can do more of 

John Willis: it. I've done a little of that where I'll do some things where I'll sort of you know, chart some data of like how people are testing software and then there's sort of some education and there's a plan, do, study, act.

Process, bring in new training and then watch it sort of, yeah, it is really interesting. And that, that sort of leads into the, the enumerated versus analytics, right? Like, which is, you know, I think the, the enumerated is, is more of the, like I apologize to any statistician who's listening, but like there's, you know, there's lies and there's statistic, like statistic lies or, you know, in our world, we, we have this thing where we call it the watermelon.

Like it's green on the outside and it's the, everything in the middle is red. Right. I love the idea. We tell narratives. Yeah, data as opposed to [00:31:00] the data giving us direction, which that's how I see analytical. 

John Dues: Yeah. And I, I definitely a debt of gratitude to the various people that, you know, were actually contemporaries of Deming or, you know, but maybe a little younger.

So they're still alive. There's a great YouTube video of a statistician named Michael Twiney. I think that's how you pronounce his last name, but it's spelled TV. The. E I T E. He's a statistician and he's got this great, you know, when I was trying to write that first chapter and I included like, you know, a description of a numerator versus analytic, it was his YouTube video that it was like three minutes.

And he broke it down in a way that, you know, one of those guys that understands him so deeply that when he explains it, it's just like, oh, okay, I got it. I got it. I got what this means. 

John Willis: Yeah. I watched that video. It was in your book, and I had seen it before. I'd forgotten, but I don't think I would ever remember this spelling.

You know, one of the things I, you know, I don't know if there's a good answer for this. I haven't asked a [00:32:00] whole lot of people this. You know, there's the rules, right? Seven rules, right? And, you know, the rule one, right? It's pretty straightforward, right? Like, you know. Common, special, you know, a special course, right?

Yeah. You know, so I tried to go back and see, because they, they're almost like, like, they're sort of, they seem to be universal, right? The thing I, I like from my understanding of a, of a control chart is that you know, and I'm paraphrasing Deming, so this is, isn't what he actually said, but it's something to the effect that it's not statistician's job to identify the problem.

It's the statistician's job to make aware to the subject matter expert so they can find the problem. To me, that's the, the, the real beauty. And the fact that those rules tend to work for anything, from life action education to software delivery, like, and so I try to go back, look at the history of like, What was the sort of original, like, rule two, right?

Like, yeah, and and I, I'm not [00:33:00] expecting, if you have the answer, that's great. But more importantly, I guess, like, you know, like, you can, over the years, I'll see that there were original Shewharts, there were some Bell Lab ones, there, there's like all these variations of the, the history of the rules. And some of them like make sense, right?

Like, but, but some of them like, okay, how did they come up with that one? Right? Like, so I think what, eight in one direction or something like that, that, that sort of makes sense. Like, you know, you can understand that something is changing in a non variant pattern. Right? But the ones that are sort of interesting is like, Like I, again, was it four or five below the line in a row or whatever that, or the, the three or four up near the top of the control?

Like, do you, have you seen examples where like, okay, yeah, I, like, we've seen that. That's a rule we need to investigate. And we found this. Yeah, 

John Dues: it's, it's a mixed bag, you know so to simplify it a little bit, you know, you referred to those seven rules. I think there's a, like a Western electric set. That's seven rules and [00:34:00] the Institute for Healthcare Improvement.

They have their own set of rules that is in some different types of software. I use Wheeler. He's just, he's like, if you use these three rules, the one you talked about rule one, where it's outside of either the limits. Rule two is eight consecutive, either above or below the average line or the central line.

And then rule three is three of four in a row that are closer to a limit than they are to the central line. I hardly ever see that one. Okay. But I basically use those first two. And if, if there's something outside the limits. You know, it's a pretty extreme difference. And so you can't always identify it, but, but, but sometimes, you know, I think it's always worth it to go study and see if you can identify it, especially if it's, you know, still close to the time that it, that it happened.

Right. Sometimes you can figure it out. Sometimes you can't, but it's worthy of study. And then, you know, the eight in a row that's sort of like a [00:35:00] moderate, but sustained indicator that either something has improved if it's in the good direction or something's. It's gotten worse. It's in the bad direction.

So I've simplified things that way. Wheeler just says, you know, these three rules will basically get you a lot of coverage, you know, So the eight 

John Willis: in a row is like a shift in something like now it was sort of variant and then all of a sudden now some shift has now caused us to have that variance. Yeah, yeah, below the line or if it's just, okay, that's what Wheeler 

John Dues: says is because.

You know, if you're trying to improve something, you know, let's, you know, make something up like my weight, if I'm trying to lose weight or something, and I'm charting my weight, 

John Willis: right. 

John Dues: And you know, let's, let's say I do see a rule one signal, right. But I, I don't really, I don't know exactly why that is. 

Wheeler says don't shift the limits, don't shift the limits, you know, but, but if you can identify, okay, [00:36:00] I get this, especially if it's eight in a row, I've been doing something new. Right. I know why it probably shifted about this time. I'm going to shift the limits in that direction. And I've kind of created a new system for whatever that thing is, you know, and so when I'm running PDSAs.

When I see one of those signals, that's when I'll, that's when I'll shift the limit, because I can tie it back to that particular PDSA. 

John Willis: Okay. Oh, yeah, yeah. That makes sense. That makes sense. And then I'm going to keep going 

John Dues: with it if it's, you know, in the direction of good. As long as I still care about improved, maybe that thing is improved enough and I'm going to turn my attention to something else.

So 

John Willis: there was another question I had, and this 1 again, maybe a little far fetched, but it was just something that stuck with me when I was I was thinking about because I, you know, I, you know, I, I don't do a lot of hands on statistical process control. I talk about it. I write about it. I try to advise people.

I try to get people who are trying it to tell me what they've done. [00:37:00] When I looked at that early chart that you have, the eighth grade is in the grades from 80 to 50, you know, 50 to 85. So my third, first thought was like, there's a lot of variance in there, and that's sort of the voice of the process. Yeah.

Sort of in, in sort of like, what do you do? Because that's like that, like looking in from the outside, not understanding anything. It doesn't look great. Right? Right. And so like, what, what, what sort of a high level process of when, when you have data like that, and you're going to try to do something with it?

John Dues: Yeah, I mean, I think it's probably not unexpected because the it's in a remote learning system. And so, you know, the wider, the wider limits to me communicate that. There is wide variability and wide variability communicates that this isn't a great processor or system. You know, we're losing quality for some reason.

So that's one thing. And so, I mean, probably what I'm going to do in a system like that is. Go [00:38:00] work with the teacher to run PDSAs to improve that system. But the thing, the thing that was like an aha moment was David, when I was working with David Lankford and the fellows, is they would bring him data like this, something like this.

And he would say, well, did you show students? And he would say, well, did you ask students what's going wrong in this system? How to improve this? He would do that over and over and over to the point where somewhere in the book, I say, like, if you want to bring about improvement in schools, you need the frontline worker.

That's the student. You need the leader. That's the teacher in a classroom and you need someone that has profound knowledge and David would say over and over you need those three things or you're never going to bring about improvement, especially the students because they're understanding why quality is being lost.

They're going to give you insights that you did not. Understand or didn't have because you're not seeing things from their perspective. So that's what I would do. I would, I [00:39:00] would, you know, if I could, I would go to that classroom and plan a with the class and I would show the class this data. 

John Willis: No, this is really interesting because I, you know, like, you know, and it probably is just my misunderstanding, but I always think of, like, when I'm looking at control trial, I'm looking for. 

Anomalies and that's not the right word, like, but I'm either looking for, you know, sort of rule one or sort of rule. Yeah, but what what you're saying is like in that scenario, which again, when I was trying to think, like, what do you do here and what you're saying is David Langford which I love his his videos or his podcast on it and on the Devin Institute, but what he's saying is like that, like, okay, there's something that's helping me improving.

I've got the right variation. I don't have any patterns. They're all within the, it is the voice process, but you're saying that that basically what that means is, and you need somebody who understands profound knowledge. You then take, you know, sort of [00:40:00] whoever is sort of leading that and the students and have them just explain it.

So there's no outright anomaly or patterns, but that in that case, you open up the discussion just like you would. Like Deming said, if, if I, if I see like eight in a row, like going up or something like that, or eight below the line, right? Like, I'm going to say, okay, let me find somebody who knows about this.

What was the time? And we're going to sort of have a conversation and it was, Oh yeah, I know we brought in a new machine or, but in any other case where there's no obvious pattern, it's still the same process. 

John Dues: Yeah, basically. So what I would say is. In that remote learning data, there's wide variation, but there's no, special causes.

So you have this stable, but unsatisfactory system. Right, right. So, so the only way to improve is, there's nothing to study. There's no, like, dominant cause to go study and move. [00:41:00] You have to make some fundamental change to the system, and I'm going to do that by running a PDSA, and I'm going to, Design that PDSA by using the people in the system, the teacher and the kids basically to do that.

John Willis: No, I love that. Yeah. I guess the, the one that like, you know, so I, you know, I've been all in on Deming for quite a few times. I love, you know, like you, you sort of fall in love with the man, you fall in love with like everything about him. Yeah. You start to learn about what a humanist he was, the kind of person he was, and then, and then the way he thought about the world, the way he was you know, he was just inquisitive, right?

Like, he seemed to want to learn everything he possibly could, you know, like, how could you not adore a person like that? Right. One area I struggle, and I had Dick Steele on my thing, and I, you know, so people who have, you know, that when I read I go back and I'll read Out of the Crisis again, and the part that sort of, sort of still makes me feel uncomfortable.

Cause I personally have never implemented a non [00:42:00] merit system. Now I understand what's wrong with the merit system. I, I like, I get all this and it, at the institutional level, absolutely. Right. The, the sort of the judgment, the fear, you know, the, the, the, the way people game, you know, like but I've had software companies where like, I need my salespeople to be the highest, like I need that.

And it went in highly competitive scenarios. And I'm sure you have sort of this. So like my son went to a magnet school, right? Like it's gotta be some teachers that are just gonna like, I, you know, and, and so I still struggle with, like, I get creating an environment where everybody win, win the name of your book. 

I, I, I get that and I get it at scale. I do, but there are certain scenarios like what I've had companies where, you You know, when you're a small startup company and you're trying to sell to JPMorgan Chase, there's a certain kind of people who could bring paper in [00:43:00] like that. And, you know, and I've had teachers that like, you know, like, so I don't know.

John Dues: Yeah, that's a good question. I mean, there's, there's no doubt by, you know, the rule is there's going to be variation between the performance of the people, whether you're talking about salespeople or teachers or athletes or whatever. So that, that's, that's the rule, not the exception for sure. Still one question would be.

Is is that performance a special cause? So if it is a special cause, then there may be a reason to do something differently with that, you know, in this case, an individual person, or you could at least say you could actually back up the claim that, yes, this person is actually being. Different their performance is different than what you would expect from this particular system.

So I guess one, you want the evidence that that's the case. And I think, you know, what I've heard is, is there another way to set things up? [00:44:00] Because like with teachers, I can't speak as well, maybe about the salespeople with teachers. The things that have been tried, there have been so many flaws, like, you know, coming out of the Obama administration, they tried to do, use value added data, basically where they would use proprietary formulas to show that students test scores grew more than expected for that same type of student in other teachers classrooms, but there's all types of problems that came, you know, the opaqueness of the, of the formula It disincentivized taking on tough teaching assignments because certain demographics of kids are harder to get results with.

You don't want to disincentivize good teachers taking on those tougher assignments. There was not, you know, all subject areas. Had state tests. There was a study I saw that it actually attributed a significant portion of this year. Like if I'm the sixth grade teachers results, a significant portion [00:45:00] of her results are actually attributable to the fifth grade and the fourth grade teacher.

So 

John Willis: these 

John Dues: systems get so crazy and do more harm than good in education at least. And so that's what I've seen. And so that's why we've kind of stayed away from any type of, 

John Willis: yeah, no. 

John Dues: or, 

John Willis: yeah. I was thinking about your whole story about the pilot and, you know, you, what makes you feel safe about flying?

Right? Like, mm, is it, and it works both ways, right? Like, so I've got a bunch of people do resilience and you know you know, sort of safety critical resilience and stuff like that. And sort of the joke is it's pilot error, right? Which is sort of a, as far, you know, like. That's why everything went wrong.

It was the pilot. You get blame or praise for both. It is not just pilot error. There's so many things that lead up to like, you know, some of these disasters, but same token. I love the way you sort of talk about, you know, why do I feel safe? Well, because I got, I think I fly airline that has great pilots.

Yeah, but great flight [00:46:00] attendants, but maybe I like the way they do maintenance, you know, which means tear the engines apart, you know, every. Six months, whatever. Right. Yeah. Yeah. So I thought that was really good. So what, what, what, what's new, what, what is the, like, is like, you know, like, I think, you know, clearly listening to you when you're reading, you're, you're sort of always trying to improve.

What's the, what are the things that you think you're finding most interesting now? 

John Dues: Yeah, that's a good question. I mean, I think, you know, When I really step back and think, you know, I, this, you know, like really got into Deming February 2020, started writing the book September 2020, and then started immediately like applying the idea.

So I'd really only been doing it, you know, for less than four, four years. So there's, you know, lots of constant learning. What I'm continuing to do, you know, within my organization is It's sort of just increasingly bringing the Deming philosophy here and it just, it takes repeated attempts and that's, that's something that I'm pretty committed [00:47:00] to.

 I don't know yet, you know, in terms of like another book or something like that. I've kicked around, you know, a few ideas, but I haven't settled on anything. So right now my main, my main focus is bringing people along at my organization into the Deming philosophy. 

John Willis: I can imagine you've gotten reached out from bigger public school systems, right, because is there, I mean, like, like, we all wish that, you know, like, I, I'm not an anti education, but I think, like, you know, my kids grew up in a school, a public school district that, See I think it was a world class education.

You know, my wife spent a lot of time trying to fit our, our school system when we, when we moved, you know, to Atlanta area, we don't live there anymore, but it's grew up with like, the teachers were incredible. So I'm, I'm very pro like, you know, but I mean, there are some, well, I'll give you an example. Like my, one of my oldest son, Was you know, when it's going to go to one of these high tech magnet schools, right.

And computer science teacher, right. [00:48:00] And, and like, it's all windows machine. And I'm like, Hey, you know, you do know you're like, what, you're supposedly one of the top technology high schools in the country. And you know, the world pretty much runs on Linux, you know, and she said, don't even get me started.

You know, and like. And she told me a little about like how the funding and what they have, you know, so, so like there are some broken things in the, in the sort of the, the public school systems. I imagine somebody is taking notice of what you've done in your system. Right? And 

John Dues: yeah, I mean, people reach out from time to time.

Actually, I I just helped. Teach a principal prep class at Drake University. The, the professor there is a long time superintendent retired now and, and helps run their ed leadership program at Drake. His name is Doug Stillwell. And so I get a lot of outreach from people like that. He was already linked into the Deming community.

And so when he found the book. You know, he [00:49:00] had had trouble finding education examples. So I'll get opportunities like that where I went in and through zoom, cause you know, Drake's in Iowa, I'm in Ohio and, and help teach a part of his class this past Saturday. So I do a number of things, you know, you know, like that to spread the word.

And then, you know, just today, someone from his class reached out and said, Hey, I'm doing this thing in my class. I don't quite understand how to use these control charts. Can you, can you send me some stuff? And so, you know, anytime someone like that's reaching out, I'm, I'm You know, helping them out however I can.

So it's a lot of, it's a lot of informal stuff like that. A lot of informal outreach, no, no full like school systems that said, like, how do we, how do we do this, you know, baby steps, I guess. 

John Willis: Yep. So the book I guess Amazon win win. 

John Dues: Yeah. Amazon's probably the best place to get it. You can get it on Meyers Education Press.

That's the, that's the publisher. 

John Willis: Yeah. And how do people reach out to you if you want them to reach out to you or? Yeah, you can 

John Dues: find me on LinkedIn [00:50:00] under John A. A. Dues. I post on there at least once a month an article and then my email is jDues at unitedschoolsnetwork. org. People can feel free to reach out there.

John Willis: I'll put some links in there too, so. All right. It was fun, man. It was good. Absolutely. That I really enjoyed your book. I, you know, I, I've gotta be honest, when it first came out and it was the reason I was waiting on, I was like, do I really wanna learn, you know, learn about the education system? But I sounded of buckled down.

I'm like, it's a great book. I mean, does it? Yeah. I appreciate 

John Dues: it. 

John Willis: Yeah. It was, I, 

John Dues: I I felt the same way about yours. I, I had tons, especially the history part, you know, when you go through. His life and how it's sort of you know, it's almost like a smart Forrest Gump type thing where he's like, that was 

John Willis: a, that 

John Dues: was unbelievable.

I really enjoyed it. 

John Willis: I needed to understand where this profound knowledge thing came from. The clarity was like, he didn't just put his finger in air and it happened. Right figure out like [00:51:00] where did those components come from? You know, how did we where did you learn it? So yeah now it's been fun learning, you know working with deming a lot of fun.

I highly recommend it So yeah, I agree. so much, buddy. It was great. So yeah, thank 

John Dues: you I appreciate it